
Distributional Reinforcement Learning is a comprehensive machine learning textbook by Marc G. Bellemare, Will Dabney, and Mark Rowland. This book introduces distributional reinforcement learning, a powerful approach to AI decision-making that models the full probability distribution of future rewards rather than relying solely on expected values.
Covering essential mathematical foundations, key algorithms, and real-world applications, it offers valuable insights into modern reinforcement learning techniques. Ideal for students, AI researchers, and machine learning enthusiasts, this free online book is an excellent resource for learning advanced reinforcement learning, probabilistic modeling, and artificial intelligence.
Table of Contents
- Introduction
- The Distribution of Returns
- Learning the Return Distribution
- Operators and Metrics
- Distributional Dynamic Programming
- Incremental Algorithms
- Control
- Statistical Functionals
- Linear Function Approximation
- Deep Reinforcement Learning
- Two Applications and a Conclusion
Download Free PDF / Read Online
Publisher: The MIT Press
Published: May 30, 2023
Format(s): PDF
File size: –
Number of pages: 384
Download / View Link(s): PDF